Learning to program a CNC lathe - FANUC Group Type A, B or C format???

Hello,

I am in the ongoing process of teaching myself to program CNC lathes (small 2 axis turning center and a tool-room lathe). I'm familiar with programming for a HAAS VMC and Milltronics knee mill, but relatively new to lathes. I've dabbled a bit with simple macros and statements to run a bar-puller and keep parts count, but nothing more advanced.


Anyway, the question is: should I be programming in the "Group Type A" G-code (it sounds like it is most common here in the USA) or does type B, or C have advantages?

I'm working with a FANUC controls and using a combination of HSMWorks (not that great for turning) and Fusion 360 (neither) and so I'll have to do some tweeking to the post-processor used by both.

jj80909 said:

Hello,

I am in the ongoing process of teaching myself to program CNC lathes (small 2 axis turning center and a tool-room lathe). I'm familiar with programming for a HAAS VMC and Milltronics knee mill, but relatively new to lathes. I've dabbled a bit with simple macros and statements to run a bar-puller and keep parts count, but nothing more advanced.


Anyway, the question is: should I be programming in the "Group Type A" G-code (it sounds like it is most common here in the USA) or does type B, or C have advantages?

I'm working with a FANUC controls and using a combination of HSMWorks (not that great for turning) and Fusion 360 (neither) and so I'll have to do some tweeking to the post-processor used by both.

Click to expand...

1st stay away from software programs for lathes its not necc unless you have some really complexed parts.
2nd learn how to read a lathe program
3rd your machine may have some other cycle codes so make sure you know them and the parms it takes to run them.
4th dont worry about a b or c until you learn to read lathe code.

lathes are stupid simple to run, you can finger bang a program out faster than you can write one up on software.

In 40+ years of Fanuc CNC experience, I have only seen a very few lathes set up with system B or C. I'd advise sticking with system A. That way if you ask for advice here, or on other forums, you'll get better answers since most folks only know system A.

For a simple 2 axis lathe, system B or C offer no advantage over A, unless you are trying get better code compatibility with another control.

Plus, I think you may have the Cart before the Horse here.

You are talking Macro. How about getting some parts off first? Once you can confidently get good parts off, then start screwing around with Macro. Everyone wants to be a genius before they can write their names. Programming a 2 Axis Turning center with CAM is retarded. You are going to waste more time screwing with the Post processor than you would think. Trying to get every detail just right, THEN you are going to go to a different Turning center and it will all be completely irrelevant. KISS.

R

litlerob1 said:

Plus, I think you may have the Cart before the Horse here.

You are talking Macro. How about getting some parts off first? Once you can confidently get good parts off, then start screwing around with Macro. Everyone wants to be a genius before they can write their names. Programming a 2 Axis Turning center with CAM is retarded. You are going to waste more time screwing with the Post processor than you would think. Trying to get every detail just right, THEN you are going to go to a different Turning center and it will all be completely irrelevant. KISS.

R

Click to expand...

Hello Rob,

He's referring to the three different G Code Systems that any one of which can be set via parameter. For example, System A uses X/U and Z/W for Absolute/Incremental respectively, whilst System B and C use G90/G91, typical of a Mill Control.

Regards,

Bill

litlerob1 said:

Plus, I think you may have the Cart before the Horse here.

You are talking Macro. How about getting some parts off first? Once you can confidently get good parts off, then start screwing around with Macro. Everyone wants to be a genius before they can write their names. Programming a 2 Axis Turning center with CAM is retarded. You are going to waste more time screwing with the Post processor than you would think. Trying to get every detail just right, THEN you are going to go to a different Turning center and it will all be completely irrelevant. KISS.

R

Click to expand...

I am so sick of hearing this. If ur making donuts, yep, just do it at the control. If you are making something other washers,pins,and donuts, learn to do it in cam. I wish I could share a print of some of the lathe parts we do at current job, or ones we did when I lived in northern IN. Good fuc*ing luck doing those at the control.

Thanks for all the replies. I'll go with keeping both lathes setup as Group Type A. I still have some work to do on the post-processor, but I ran +500 parts over the weekend with a bar-puller and parts counter. It's a nice feeling to load a bar, push a button and make $$$

Mike1974 said:

I am so sick of hearing this. If ur making donuts, yep, just do it at the control. If you are making something other washers,pins,and donuts, learn to do it in cam. I wish I could share a print of some of the lathe parts we do at current job, or ones we did when I lived in northern IN. Good fuc*ing luck doing those at the control.

Click to expand...

Yeah, but you have zero frame of reference. I'm a guy who sets, programs and operates, 9 Axis Turning centers (not Swiss). And have set, programmed and operated everything between. I use 3 different CAM suites, and 2 CAD suites, I was around when IBM handed Catia off to Boeing. I cannot imagine what complexity of parts come off a 2 Axis Lathe that require CAM. It's 2 Axes, not C, not Y. It's Turning. What you are proposing is like using CAM for a Manual Lathe.

Fly your BS flag elsewhere my friend.

R

And I promise you that for ANY 2 Axis part, I can get a good one off before ANY CAM user can.

I agree, for 2 axis lathe work you won’t need CAM 99% of the time.

litlerob1 said:

Yeah, but you have zero frame of reference. I'm a guy who sets, programs and operates, 9 Axis Turning centers (not Swiss). And have set, programmed and operated everything between. I use 3 different CAM suites, and 2 CAD suites, I was around when IBM handed Catia off to Boeing. I cannot imagine what complexity of parts come off a 2 Axis Lathe that require CAM. It's 2 Axes, not C, not Y. It's Turning. What you are proposing is like using CAM for a Manual Lathe.

Fly your BS flag elsewhere my friend.

R

And I promise you that for ANY 2 Axis part, I can get a good one off before ANY CAM user can.

Click to expand...

OK tough guy. You have a part with a tapered bore say 6-8" long, large end lets say 2"and small end .50" You have internal grooves perpindular to the tapered bore. Sure the taper is easy. How would you use a canned cycle to generate internal grooves on an angle? OK maybe you come up with some fancy macro (I don't know if it's possible or not) BUT these grooves are not all the same depth, or on an equal spacing, so maybe 1 macro for each one, I dunno. AND since every 3 or 4 of these we cut we have to make a custom groove tool for, in cad. Now we draw the tool, and they get stupid lxd ratios so we make them as large as possible, but we also need to verify (see the cam part there??) the bar will clear everything....

Maybe you could do them as we all know your superman on the lathe, but here we use cam to do all that work.

edit: just looked at the last one, it is approx 4.5" deep, 1.6" large dia, .25 small dia (thru hole), roughly 40 grooves about .08" wide, varying depths

There are occasionally parts that CAM is handy for on a 2 axis lathe. Mike presents a pretty good one though the bore being a taper makes it not too tough to calculate the intersections. If the bore was a radius, then CAM would be really handy. IME, the vast majority of 2 axis lathe parts CAM is a waste of time.

Mike1974 said:

You have internal grooves perpindular to the tapered bore.

Click to expand...

What application is it for?

Vancbiker said:

There are occasionally parts that CAM is handy for on a 2 axis lathe. Mike presents a pretty good one though the bore being a taper makes it not too tough to calculate the intersections. If the bore was a radius, then CAM would be really handy. IME, the vast majority of 2 axis lathe parts CAM is a waste of time.

Click to expand...

We have prints, and we could dim everything if needed, but still not sure if it is possible with a canned cycle or not as you would be grooving on an angle. Not to mention typing all the crap in on a machine control .. maybe there is a way.. I don't know what other controls offer for canned cycle, so maybe a limitation in Fanuc, but not Siemens or Okuma, I don't know.

But really my whole issue is "you never never need cam for a 2 axis lathe part" as a blanket statement is pretty brazen IMO. I also made some parts that all the internal features were tapers into rads with nothing 'square' except the OD and back face. The front face was all contoured. We had a guy finger camming those but always seemed to have an oops there and an oops here, which was especially troubling because there wasn't a good way to check in the machine... That and having prints using 2 place decimals (corners rounded to .13 instead of .125 and tapers as 1.2 instead of 1.204 etc) produced a lot of errors in form when typing those dims in "as is".

sinha said:

What application is it for?

Click to expand...

I don't know, we machine it and it goes into the lab for various electronic gizmos and testing then to the customer.

Mike1974 said:

We have prints, and we could dim everything if needed, but still not sure if it is possible with a canned cycle or not as you would be grooving on an angle. Not to mention typing all the crap in on a machine control .. maybe there is a way.. I don't know what other controls offer for canned cycle, so maybe a limitation in Fanuc, but not Siemens or Okuma, I don't know.

Click to expand...

I think I see the mis-communication. Ass-U-Meing that you are bound to canned cycles, then yes that poses a problem. I am not bound to them. A canned cycle is just a refined Macro anyway.

Point being that I could write that out long hand pretty damn fast. BUT I concede; in that example I would probably use CAM. There's your 1% of the time.

R

Mike1974 said:

I don't know, we machine it and it goes into the lab for various electronic gizmos and testing then to the customer.

Click to expand...

Requires pecking at an angle, with a special tool?

sinha said:

Requires pecking at an angle, with a special tool?

Click to expand...

Yes, peck moves are XZ with custom ground tools. We usually wireburn the shape of the tool, then grind clearance angle as needed.

Imagine the tool movement like this -

\
\
\

But at a steeper angle usually...

litlerob1 said:

I think I see the mis-communication. Ass-U-Meing that you are bound to canned cycles, then yes that poses a problem. I am not bound to them. A canned cycle is just a refined Macro anyway.

Point being that I could write that out long hand pretty damn fast. BUT I concede; in that example I would probably use CAM. There's your 1% of the time.

R

Click to expand...

Apology accepted! hahah (it's a joke relax)

I don't understand what you mean by you are not "bound to canned cycles"?? So it is then using conversational, or like a dxf import, or..? BTW even though I don;t like conversational in general, it is still more or less a version of cam, just specific to the machine and control.... OR do you truly mean longhand with every single X and Z move? That (writing longhand) doesn't really sound very fast or efficient to me considering these are all peck grooved with multiple "stop points" in the program to clear chips and check the tool. Now I couldn't say if they need to be peck grooved or not, might find a perfect combo of speed and feed to just plunge them which would drastically reduce the code....

At the current job we do alot of dead simple turning that the lathe guy does do at the machine. But that is just circling back to my point about simple shapes...

Maybe I have had the good fortune (take that how you want) to see lots of lathe stuff that is not so simple and justifies using cam, I dunno... I remember when we hired a lathe guy years ago, I was still new to cnc and programming at the time. He insisted it was sooo fast to do at the machine using canned cycles (and some of it is no denying). So he kind of did his thing for a while, and he was pretty damn good no doubt. Made very very little scrap and was fast about it.

Well fast forward a bit and "hey can you get me this endpoint, this endpoint, intersection here". No problem. "this ain't right, keep getting an alarm at machine" Show him, here is what I picked and here are the numbers, it was rounding errors on the prints causing the issue. So after learning what depth of cut (I was pretty green on lathe stuff in general), what speeds/feeds, what tooling/rads/loc-depths we had to work with. I started programming that stuff in a fraction of the time and even he was impressed. Point not being me, but when you have cam available and you can pick a profile and tool, doc, etc it beats the hell out of typing all those numbers! He even learned enough of the cam side so when we got those kind of parts he could just sit down and do it without monkeying around getting points and stuff...

Why would anyone want to use canned cycles for something simple like grooves? It is not anymore difficult to hand write a groove than straight OD turning or ID boring.
Even in the case of the grooves perpendicular to the taper, as long as the tool is ground to the same angle of the taper and you know where you touched off the X and Z you can figure out the moves that it needs to make.
I understand everyone is different and learn and do things differently. I just will never use canned cycles for grooves.

@ Mike; I am very fast.

I don't mean exclusively at hand coding Machinery (though that's where the need to be, started) I mean, I am a Godamned whirlwind. I make very little scrap, usually because I need to. Using an OSP as an example, I would think just from your description of the part in post #10...I think I could have it coded in less than an hour. It would in fact take less time for me to do it in Esprit or Mastersuc, but there is considerable peripheral time with using CAM as you know. So I would think it would be comparing Keystone light to Natural light. Not even worth the argument.

Point being that I spent years and years hand coding everything in a Job-Job shop. So I got fast.

The only thing I use canned cycles for (on a 2x Lathe) is Drilling, and occasionally Rough Turning. The rest of the time it's long hand G0-G1-G2, but like I said you spend a few years doing it under the gun and you get fast. With all that being said, I do build and prefer to build the parts that no one else in the world wants to build. I get off on the toughest jobs around.

R